Github user lokkju commented on the issue:
https://github.com/apache/spark/pull/16479
So it turns out just copying the conversion code doesn't work, as seen in
spark-avro/#240 - and now I'm running into the same thing writing my own
datasource. As an datasource in the end requires implementing a class that
extends OutputWriter, and the OutputWriter interface changed, a datasource
plugin doesn't seem to be able to support both pre and post versions of 2.2.x
in the same plugin.
Any suggestions on how to handle this, without requiring users to match a
the spark version to the new datasource version?
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
